Comments (1)
Thanks for opening this, @alegarbed! Yes, you can optimize class_weight
values! Here's a basic example with SKLearn's RandomForestClassifier
and the Iris dataset.
from hyperparameter_hunter import Environment, CVExperiment
from hyperparameter_hunter import BayesianOptPro, Integer, Categorical
from hyperparameter_hunter.utils.learning_utils import get_iris_data
from sklearn.ensemble import RandomForestClassifier
env = Environment(
train_dataset=get_iris_data(),
results_path="HyperparameterHunterAssets",
target_column="species",
metrics=["hamming_loss"],
cv_params=dict(n_splits=5, random_state=32),
)
# Just a reference for normal `class_weight` usage outside of optimization
exp = CVExperiment(
RandomForestClassifier, {"n_estimators": 10, "class_weight": {0: 1, 1: 1, 2: 1}}
)
opt = BayesianOptPro(iterations=10, random_state=32)
opt.forge_experiment(
model_initializer=RandomForestClassifier,
model_init_params=dict(
#################### LOOK DOWN ####################
class_weight={
0: Categorical([1, 3]),
1: Categorical([1, 4]),
2: Integer(1, 9), # You can also use `Integer` for low/high ranges
},
#################### LOOK UP ####################
criterion=Categorical(["gini", "entropy"]),
n_estimators=Integer(5, 100),
),
)
opt.go()
This should definitely be included in one of our examples, or at least documented, so thanks for asking again!
Side note: I just noticed that the automatic Experiment matching during optimization isn't working for this, which is a bug, so I'll look into that and update you
from hyperparameter_hunter.
Related Issues (20)
- Support for various advanced functionality? HOT 9
- Unable to access docs HOT 2
- How to do predict_proba in catboost classifier? HOT 3
- Support for nested parameters/parameterizing objects that can't be called by name. HOT 2
- How to handle variable number of layers?
- How to use the Experiments? HOT 6
- OSError: could not get source code HOT 9
- Q: what is RandomForestOptPro exactly? HOT 3
- Doc improvement suggestion HOT 4
- logloss issue with multiclass task HOT 4
- Metaclass conflict with Keras 2.3.0 HOT 1
- Timeseries Cross Validation HOT 6
- Any way to send hyperparameter_hunter.Integer to custom feature_engineer function? HOT 3
- pd.DataFrame.sparse supported in environment?
- Interface with pytorch or tensorflow models HOT 2
- Problem with library scikit-optimize in Python 3.6 HOT 9
- ROC AUC scores don't match to those of sklearn HOT 1
- Failed to import packages from hyperparameter_hunter HOT 3
- ImportError: cannot import name 'Log10' HOT 1
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from hyperparameter_hunter.